Sony's Wireless Power Transfer can push 60 watts of electrical energy over almost 20 inches (50 centimeters). That's a pretty decent distance, especially when they say that it can be extended using passive extender units. In fact, they have already achieved 31 inches in other tests.

They claim that their method—which sounds similar to Intel's—uses some dharmastastic magnetic resonance, in which electromagnetic energy gets transmitted from one device to another, both sharing the same resonant frequencies.

Sony says that this system offers 80% efficiency, which may get reduced to about a minimum of 60% if there is a misalignment in the frequencies, which needs to be corrected.

I don't know if this is safe for our bodies or not, but I would like to have it. Even if that requires me not wearing my tinfoil hat while watching the sixth season of Lost. [Sony via i4u]

Broadcasters pump out tons of of RF from their big microwave towers, operating on the mere hope that some of the RF will hit a TV antenna and deliver unto someone the evening news. Since power demands for electronic devices continues to reduce (see Moore's Law), those radio waves can now act as currents in a stream, turning the digital wheels inside small electronic devices. The catch is that the antenna harvesting the electricity has to be in line-of-sight with the microwave tower. On the bright side, the TV station (or cell tower or home Wi-Fi network) will never feel the burden of these added devices. It's just RF that didn't make it to its intended location.

The same team at Intel Labs Seattle also figured out a way to develop motion-sensing RFID tags that require the same off-the-shelf RFID transceiver used to simply count boxes and other simple tag apps—in other words, gear that's already in place in many buildings. By sticking the little tags on a bunch of household products in a room, the researchers could track what people were doing with 90% accuracy. Some people are already testing these Wireless Identification and Sensing Platform (WISP) RFID chips for use inside the human body (pacemaker location) and deep under the sea (testing seawater 1km below the surface).

The thing is, none of these technologies are going to charge your phone or power your laptop. For that, you'll need Intel's other wireless power initiative, Wireless Resonant Energy Link, first shown off in 2007. Currently, a demo model features a 45W lightbulb operating at full brightness at 1 meter with around 80% efficiency. And best of all, it doesn't electrocute people when they walk by. [Intel Labs Seattle]

During the presentation, they provide some background on Nikola Tesla's experiments and attempt to recreate them by powering a light bulb wirelessly at various distances. It works of course—in 1899 Tesla managed to transmit 100 million volts of power over a distance of 26 miles where it lit up a bank of 200 light bulbs and an electric motor. So why haven't figured out how to do this on a large scale over the last 100 years? [Fora]

The most interesting part of Rattner's presentation was his discussion of power harvesting, and how they're looking at ways to incorporate free, renewable energy into personal devices as part of a hybrid power architecture. He thinks that technologies pulling power from light, but also heat, movement (such as rolling a Blackberry trackball), and wireless power transmissions have progressed to the point to where they can make personal devices markedly more efficient.

However, at this point, it's not efficient enough to power a gadget alone, but with a combination of adaptive power, intelligent power management, the traditional battery will last much longer without a charge.

When asked about the possibility of power harvesting from sonic sources, like the supposed piezoelectric phone concept which picks up a charge from talking, Rattner indirectly dismissed it, saying Intel had explored it back in the past, but were not currently researching the technology.

In any case, Rattner thinks it's possible for power harvesting to appear in consumer devices in as little as two years, provided they can prove it works and develop a spec for it. He says it's much easier to push new innovations into the market when they're not directly tied to a processor (those normally take 4-5 years).

Other projects he mentioned as part of Intel's research include servers with hardware-based management to better monitor the drops and increases in needed power, mobile sensors that can be used to monitor air quality on the fly and sensors which can power themselves, reporting back whatever information is needed.